University of Arkansas, FayettevilleScholarWorks@UARK

School Choice Demonstration Project Education Reform

1-1-2006

An Evaluation of the effects of DC's voucherprogram on public school achievement and racialintegration after one yearJay P. GreeneUniversity of Arkansas, Fayetteville

Marcus A. WintersUniversity of Arkansas, Fayetteville

Follow this and additional works at: https://scholarworks.uark.edu/scdp

Part of the Education Policy Commons, Public Policy Commons, and the Social PolicyCommons

This Article is brought to you for free and open access by the Education Reform at ScholarWorks@UARK. It has been accepted for inclusion in SchoolChoice Demonstration Project by an authorized administrator of ScholarWorks@UARK. For more information, please contact scholar@uark.edu,ccmiddle@uark.edu.

Recommended CitationGreene, Jay P. and Winters, Marcus A., "An Evaluation of the effects of DC's voucher program on public school achievement and racialintegration after one year" (2006). School Choice Demonstration Project. 26.https://scholarworks.uark.edu/scdp/26

An Evaluation of the Effectsof D.C.’s Voucher Program on

Public School Achievement andRacial Integration After One Year

Jay P. Greene, Ph.D.Endowed Chair and Head of the Department of Education Reform

University of ArkanasasSenior Fellow, Manhattan Institute for Policy Research

Marcus A. WintersSenior Research Associate, Manhattan Institute for Policy Research

Doctoral Academy Fellow, University of Arkansas

No. 10 January 2006

EducationWorking Paper

C E N T E R F O R C I V I C I N N O V A T I O NAAAAA T T H E M A N H AT T H E M A N H AT T H E M A N H AT T H E M A N H AT T H E M A N H A T TT TT TT TT T A N I N S T I T U T EA N I N S T I T U T EA N I N S T I T U T EA N I N S T I T U T EA N I N S T I T U T E

C C i

Education Working Paper 10 and SCDP-06-01

January 2006

An Evaluation of the Effects of D.C.’s Voucher Program onPublic School Achievement and Racial Integration After One Year

January 2006

EXECUTIVE SUMMARY

This study evaluates the initial effect Washington D.C.'s Opportunity Scholarship Program (OPS) on theacademic performance of public schools and its effects on the opportunities District students have to attendintegrated schools. OPS is a federally sponsored school voucher program that provides vouchers worth upto $7,500 for an estimated 1,800 to 2,000 students in the District of Columbia. Students can use the scholar-ships to pay tuition at participating private schools in the District. The pilot program is designed to last for5 years.

The authors measure whether a public school's test score gains are related to its distance to the nearestvoucher accepting private school or the number of voucher schools within a one mile radius of a publicschool. In theory, public schools with shorter distances to private schools or that have more private schoolsnearby should face greater competition from the voucher program than public schools with fewer educa-tional alternatives.

The evaluation finds that the OPS has had no academic effect, positive or negative, on the District's publicschools after its first year. This finding is different than most other studies, which tend to indicate schoolchoice programs have helped to improve public school performance. The authors argue that a null findingcould be explained by the fact that the OPS program was designed to have a minimal financial impact onpublic schools. They also suggest that the null finding could be explained by the small size of the program,the short time-span in which it has operated (1 year), methodological considerations, or a true lack of arelationship between vouchers and academic performance in Washington D.C.

The paper also compares rates of racial integration in D.C.'s public schools and private schools participat-ing in the voucher program. The authors find that voucher accepting private schools have populationswhose racial demographics more accurately mirror those of the surrounding metropolitan region than dopublic schools in the District. The study also finds that students using an Opportunity Scholarship are lesslikely to be enrolled in a school that is 90% or 95% racially homogeneous than are students attendingWashington D.C. public schools. This finding, combined with a previous evaluation indicating that the vastmajority of students participating in the OPS are African-American, suggests that OPS will likely lead tostudents leaving more segregated public schools for better integrated private schools.

This is part of the first year evaluation of the OPS program. The authors plan to continue evaluating theOPS program using a variety of approaches.

Education Working Paper 10 and SCDP-06-01

January 2006

ABOUT THE AUTHORS

Jay P. Greene is Endowed Chair and Head of the Department of Education Reform at the University ofArkansas, and a Senior Fellow at the Manhattan Institute. He has conducted evaluations of school choiceand accountability programs in Florida, Charlotte, Milwaukee, Cleveland, and San Antonio. He has alsorecently published research on high school graduation rates, social promotion, and special education.

His research was cited four times in the Supreme Court’s opinions in the landmark Zelman v. Simmons-Harris case on school vouchers. His articles have appeared in policy journals, such as The Public Interest,City Journal, and Education Next, in academic journals, such as the Teachers College Record, the GeorgetownPublic Policy Review, and the British Journal of Political Science, as well as in major newspapers, such as theWall Street Journal, the Washington Post, and USA Today.

Greene received his B.A. in history from Tufts University in 1988 and his Ph.D. from the GovernmentDepartment at Harvard University in 1995. He lives with his wife and three children in Fayetteville,Arkansas.

Marcus A. Winters is a Senior Research Associate at the Manhattan Institute, where he studies and writeson education policy. He is also a Doctoral Academy Fellow in the Department of Education Reform at theUniversity of Arkansas. He has performed several studies on a variety of education policy issues includ-ing high-stakes testing, charter schools, and the effects of vouchers on the public school system. His op-ed articles have appeared in numerous newspapers, including the Washington Post, USA Today, and theChicago Sun-Times. He received his B.A. in political science with departmental honors from Ohio Univer-sity in 2002.

ACKNOWLEDGMENTS

We would like to thank Patrick Wolf for his help on this project.

ABOUT EDUCATION WORKING PAPERS

A working paper is a common way for academic researchers to make the results of their studies available toothers as early as possible. This allows other academics and the public to benefit from having the researchavailable without unnecessary delay. Working papers are often submitted to peer-reviewed academic journalsfor later publication.

An Evaluation of the Effects of D.C.’s Voucher Program onPublic School Achievement and Racial Integration After One Year

January 2006

TABLE OF CONTENTS

Previous Research and Theoretical Expectations on the Effect of Vouchers

on Public School Academic Performance............................................................................................1

Previous Research and Theory on the Effects of

Vouchers on Racial Integration ............................................................................................................3

Evaluating the Effect of Vouchers on D.C. Public Schools After One Year ................................................5

Results and Discussion of the Effect of Voucher Competition on

Public School Academic Performance .................................................................................................6

Evaluating the Potential Effects of Vouchers on Racial Integration ...........................................................7

Results and Discussion of Evaluation of Racial Integration in Public and Voucher Schools .......................8

Conclusion .................................................................................................................................................9

Endnotes ..................................................................................................................................................10

References ...............................................................................................................................................10

Appendix .................................................................................................................................................12

Table 1: Effect of Voucher Competition on Public School Performance ...........................................12

Table 2: Measures of School Segregation .........................................................................................12

Figure 1: Percent Minority Enrollment in Deciles ..............................................................................13

Education Working Paper 10 and SCDP-06-01

January 2006

January 2006 1

AN EVALUATION OF THE EFFECTS OF

D.C.’S VOUCHER PROGRAM ON

PUBLIC SCHOOL ACHIEVEMENT AND

RACIAL INTEGRATION AFTER ONE YEAR

In 2004, the United States Congress implemented thefirst federally sponsored school voucher programright in its own backyard. The DC School Choice In-centive Act provides vouchers worth up to $7,500 foran estimated 1,800 to 2,000 students in the District ofColumbia. Students can use the scholarships to paytuition at participating private schools in the District.The pilot program is designed to last for five years.

The existence of this pilot program offers an impor-tant opportunity to learn more about the effects ofexpanded school choice on the performance of stu-dents who exercise choice, the performance of stu-dents who remain in traditional public schools, theopportunities for students to attend racially inte-grated schools, and other community effects. Thispaper focuses on the “systemic effects” of the pro-gram: its effects on the performance of students whoremain in traditional public schools and its effectson opportunities for integration in school.

This paper examines the D.C. program after only asingle year of implementation. Therefore, the studyis limited in several ways, and later studies evaluat-ing the effects of the program in years to come mighthave substantially different results. Nevertheless, itis important to follow the progress of this congres-sionally mandated program throughout its imple-mentation in order to keep policymakers and thepublic up-to-date on its consequences.

Previous Research and TheoreticalExpectations on the Effect of Vouchers onPublic School Academic Performance

Congress implemented the voucher program becauseof the near-universal understanding that the publicschools in the District of Columbia are not living upto their expectations. As of the 2001–02 school year,the most recent year for which data are available fromthe National Center for Education Statistics, Wash-ington, D.C., public schools spent $15,489 per pupil—

substantially more than any other state1 (the next-high-est state was New Jersey, at just under $13,000 perpupil). Despite such a high funding level, the Districtconsistently ranks among the bottom of the nation ineducational outcomes. For instance, on the 2005 ad-ministration of the National Assessment of Educa-tional Progress, 58% of the District’s African-Americanstudents scored below the Basic benchmark on theeighth-grade reading test. The District’s performancewas significantly worse than the already alarmingnational average of 49% of African-American studentsscoring below Basic.2 A study by the Manhattan Insti-tute found that the District’s performance ranked lastin the nation on a previous administration of NAEP,even when the larger than average demographic andeconomic challenges of its students are taken into ac-count (Greene and Forster 2004).

The District’s persistent poor academic outcomeshave inspired a series of educational reforms. Wash-ington, D.C., was one of many school systems to haveimplemented a high-stakes accountability testingprogram before the No Child Left Behind Act madethem universal. The District has also experimentedwith school choice: as of the 2004–05 school year,there were 34 charter schools operating in the Dis-trict of Columbia.3

The federally sponsored school voucher program isthe latest attempt to provide students in Washing-ton, D.C., with a higher-quality education. The theorybehind the program is that parental choice in educa-tion should improve student learning, both for thosewho actively choose and for those who remain intraditional public schools. Whether that theorymatches the experience with the program is the pri-mary purpose of this evaluation.

Supporters of expanded school choice have arguedthat voucher programs not only help students whouse the scholarships, but lead to significant improve-ments in the performance of nearby public schools

Education Working Paper 10 and SCDP-06-01

January 20062

as well. The idea is that vouchers might providepublic schools with an incentive to improve theirperformance by increasing market competition ineducation. When students have the financial powerto leave public schools that are not serving them well,it may be harder for those schools to take studentsand the revenue that those students generate forgranted. Public schools that improved their perfor-mance would be better able to retain or even attractstudents and revenue.

Opponents of expanded school choice argue thatvouchers will harm public school performance bydepriving them of important resources. In mostvoucher programs, when a student uses a voucherto leave a public school, that public school no longerreceives the per-pupil funding that it previously re-ceived for educating the student. This loss of fund-ing could leave already struggling schools with fewerresources, which in turn could cause them to fall fur-ther behind. And if voucher programs attract themost capable students and the most active families,public schools will lose these catalysts for improve-ment and positive peer influence, further hinderingtheir ability to improve.

There is a wide and growing body of research on theeffects that vouchers and other school choice pro-grams have had on the academic performance of tra-ditional public schools. Researchers have utilized avariety of strategies to study the systemic effects ofexisting school choice programs across the nation.

There have been four empirical evaluations of theeffect of Florida’s Opportunity Scholarship vouch-ers on low-performing public schools in the state.The statewide program provides tuition scholarshipsfor students enrolled in public schools that earn twofailing grades within a four-year period under thestate’s accountability system. Independent evalua-tions of the program by Greene and Winters (2004),Chakrabarti (2005), West and Peterson (2005), andFiglio and Rouse (2005) all found that the programhas improved the performance of surrounding pub-lic schools. While Figlio and Rouse raised doubtsabout how much of the improved performance couldbe attributed to competitive pressure as opposed toa “stigma effect,” the other three studies conductedadditional analyses that led them to conclude thatexpanded choice and competition were largely re-sponsible for the gains.

DRAFT AS OFDEC 8, 2003

11:15 am

CONFIDENTIAL -- DO NOT CIRCULATE

Researchers have also paid close attention to thepublic school effects of other publicly sponsoredvoucher programs. Hoxby (2001) and Greene andForster (2002) found that Milwaukee’s voucher pro-gram substantially improved the city’s publicschools. Looking at different stretches of time, bothstudies found that public schools exposed to greatercompetition from the voucher program, by virtueof having more of their students eligible to partici-pate, made greater gains on achievement tests.

Hammons (2001) evaluated the impact of century-old voucher programs in Maine and Vermont,known locally as “tuitioning,” in which some com-munities never built public high schools and insteadoffered families vouchers to pay tuition at privateor other public schools. Hammons found that pub-lic high schools closer to tuitioning areas, and thusfacing greater competition from nearby public andprivate schools in their efforts to attract tuitioningstudents, had significantly higher test-score perfor-mance than other public schools in those states.

Other research has focused on the effect of privateschool competition on public school performancemore generally, without the use of school vouchers.Jepson (1999) and Sander (1999) each found no ef-fect from general private school competition onpublic schools. However, Hoxby (1994) and Dee(1998) both found statistically significant and sub-stantial positive effects from private schools on pub-lic school performance.

There exists a much larger body of research on theeffects of school choice between public school dis-tricts on school performance. In theory, residentialchoice between school districts, often referred to asTiebout choice, is greater where there are more pub-lic school districts operating within a reasonableproximity to one another. Where districts are morenumerous, it is easier for parents to move from onedistrict to another if they are dissatisfied with theircurrent public school. This greater residential choicemight lead to greater competition for students be-tween school districts, which could improve publicschool performance in the same way that theorysuggests vouchers could lead to improvements.

Most of the research on the effect of Tiebout choiceon public school performance has produced dis-tinctly positive results (see Borland and Howson

An Evaluation of the Effects of D.C.’s Voucher Program onPublic School Achievement and Racial Integration After One Year

January 2006 3

1993, Zanzig 1997, Hanushek and Rivkin 2002, Blairand Staley 1995, Greene 2002, Marlow 1997, Hoxby2000, and Walberg 1993). There are some studies,however, that have produced more mixed findings(Borland and Howson 1992, 1995; Marlow 1999), butnone of these findings was distinctly negative for theuse of Tiebout choice.

In a survey of the existing research, ColumbiaUniversity’s Belfield and Levin (2002) concluded thatthe culmination of the research suggests that schoolchoice likely has a modestly positive effect on theeducational outcomes of public schools. While thereis certainly room for more research on the effects ofvouchers on public schools, so far the evidence tendsto support the theory that public schools improvetheir performance in response to expanded choiceand competition.

However, there are important differences betweenprevious voucher programs and the DC SchoolChoice Incentive Act that might lead to substantiallydifferent results. Only a limited number of childrenin the District are able to use the vouchers to leavetheir public school and attend a private school. Bydesign, the pilot voucher program can only providescholarships to an estimated 1,800 to 2,000 of theroughly 76,000 students in the D.C. public schoolsystem. The limited size of the program should re-duce our expectations about the systemic effects ofthe D.C. voucher program, for good or for ill.

More significant, the D.C. choice program was spe-cifically designed to hold the public school systemfinancially harmless for the loss of students to thevoucher program. Congress explicitly declared itsintention that the choice program ought to have nonegative financial impact on D.C. public schools,writing into the text of the law: “This title providesadditional money for the District of Columbia pub-lic schools and therefore money for scholarships isnot being taken out of money that would otherwisego to the District of Columbia public schools.”4

Theoretically, holding the public school system fi-nancially harmless under the voucher program couldseverally limit any systemic effects of the policy, posi-tive or negative. The theoretical benefit of schoolchoice policies on public schools comes directly fromthe increased financial incentive that potentially los-ing enrollment funds provides. On the other hand,

theoretical concerns about how losing revenuewould hinder school improvement would also belargely rendered moot if the system faced no loss ofrevenue from the program.

Even if all the financial repercussions of a voucherprogram are removed, however, it is possible thatschool choice might affect public school perfor-mance, either negatively or positively. For example,one could argue that even schools that are held harm-less against the loss of revenue as they lose studentsto a voucher program might still feel increased pres-sure to improve in order to minimize the politicalembarrassment caused by an exodus of students.Schools might even anticipate that they would notbe held harmless against financial losses forever andbe motivated by that prospect of declining revenuein the future. Furthermore, when schools are heldfinancially harmless for losing students, their per-pupil expenditures necessarily increase as the samenumber of dollars are used to educate fewer stu-dents. We might also expect their class sizes to de-crease as students use vouchers, since the samedollars are available to hire the same number of stu-dents to educate fewer students. As long as theseextra resources per pupil are used effectively—astrong assumption—we might expect a choice pro-gram in which schools lose students without losingmoney to improve the achievement of students.

On the other hand, the loss of enrollment even with-out the loss in their funds might demoralize theirstaff, resulting in decreased student performance. Inaddition, if only the best and brightest students withthe most involved parents use the vouchers, schoolperformance might suffer because lower-perform-ing students will no longer have these exceptionalstudents as role models, and schools will lose thesupport of their most valuable parental resources.Thus, while holding public schools financially harm-less from the voucher program could significantlyaffect potential systemic responses to the program,the magnitude or direction of this bias is unclearwithout empirical evaluation.

Previous Research and Theory on the Effects ofVouchers on Racial Integration

Another outcome of school vouchers considered bythis paper is the effect they might have on the op-portunities that students have to attend a racially

Education Working Paper 10 and SCDP-06-01

January 20064

integrated school. Schools are expected to do morethan convey academic skills. We also look to them tohelp in the development of future generations of citi-zens. The positive experience with people from dif-ferent backgrounds resulting from racial integrationis another important aspect of whether schools areserving public purposes.

In particular, expanding school choice raises con-cerns about this public purpose of education. Offer-ing vouchers to attend racially segregated privateacademies was one of the strategies used by South-ern segregationists to evade efforts to integrate pub-lic schools. This negative historical association alarmssome that current voucher programs may be simi-larly motivated or have similar consequences. But itis also the case that public schools were segregatedby law in much of the country for most of their his-tory, so public schools also carry negative associa-tions as far as segregation goes. In the end, we haveto judge the effect of expanded school choice inschool integration by its effects and not by its pedi-gree.

There are also some theoretical reasons to expect thatexpanding school choice ought to improve schoolintegration. Most public schools assign students toschools based on where students live. By attachingschooling to housing, public schools may replicateand even reinforce racially segregated housing pat-terns. Vouchers may diminish this connection be-tween racially segregated housing and raciallysegregated schools by making it easier for studentsto attend schools outside of their attendance zonesor district. On the other hand, some may have theo-retical expectations that expanding school choicewould exacerbate segregation in schools by facili-tating families to act upon racist inclinations andselect schools that were even more racially segre-gated than the ones to which they were assigned.

There have been several studies comparing rates ofracial segregation in public and private schools,much of which purports to find that private schoolsare more racially segregated than public schools.However, most of this research fails to properly de-fine racial integration, leading to improper conclu-sions (Greene, 2005).

Some studies define greater integration as schoolswith larger numbers of minority students, others as

evenness in the distribution of students amongschools within already segregated school districts.Some researchers have used levels of racial integra-tion of public schools as the benchmark to measurethe racial integration in schools of choice. Finally,researchers studying the effect of school choice pro-grams, such as vouchers, have sometimes wronglycompared the demographic characteristics of thosewho participate in the programs with those whochoose not to participate as an indication of the ef-fects of the programs on school integration.

Each of the methods to measure racial integrationdescribed above fails to square with the commonunderstanding of racial integration. If larger num-bers of minority students were a proper indicatorof greater racial integration, then African-Ameri-can schools of the Jim Crow era were perfectly in-tegrated. Evenly distributing racial populationsamong schools in a district is no achievement forracial integration if an all-black school district isgeographically adjacent to an all-white school dis-trict. Each district could perfectly distribute its ra-cially homogenous student population across theschools within its district and still fail utterly to of-fer an integrated school environment. Using publicschools as a benchmark for perfect integration isalso a flawed method, considering that it means bydefinition that there is no possibility for choiceschools to be more racially integrated than the pub-lic schools against which they are being compared.And comparing the characteristics of those whochoose to participate in school choice programs withthose who choose to remain in their public schoolconfuses differences in who participates with effectson integration. A magnet program that largelydraws white students to attend predominantly Af-rican-American schools could enhance integrationeven if—or perhaps precisely because—it differen-tially attracted white students.

A more reasonable approach to measuring racial in-tegration involves comparing the demographic char-acteristics of schools with those of their surroundingmetro area. To the extent that schools contain a ra-cial mix of students that more closely resembles theracial mix of students in the broader community fromwhich they could reasonable draw students, giventransportation constraints but ignoring politicalboundaries such as city or school-district line, thebetter integrated they are.

An Evaluation of the Effects of D.C.’s Voucher Program onPublic School Achievement and Racial Integration After One Year

January 2006 5

Another reasonable approach to measuring integra-tion (or the lack of it—segregation) is to see howmany schools are racially homogenous. For instance,a school with a population that is more than 90%minority cannot be considered to be racially inte-grated under any reasonable standard. If a large per-centage of an area’s schools are more than 90%homogeneous, we could reasonably consider thoseschools to be racially segregated.

Greene (1998) examined data from nationally repre-sentative samples of public and private school stu-dents collected by the U.S. Department ofEducation’s National Educational LongitudinalStudy. Greene found that twelfth-grade students inprivate school classrooms had racial compositionsthat were on average closer to national racial demo-graphic characteristics than students in public schooltwelfth-grade classrooms. He also found that privateschool students were significantly less likely to be inclassrooms that were more than 90% racially homo-geneous than were their public school counterparts.

Ritter, Rush, and Rush (2002) replicated Greene’smethod but looked at racial segregation among kin-dergarten students rather than twelfth-graders. Theyfound that private school kindergartens are more ra-cially segregated than public school kindergartens.However, it is likely that twelfth-grade enrollmentstell us more about racial segregation than informa-tion on kindergarten students. Unlike high school,full-day kindergarten is not offered in all communi-ties, causing a significant number of wealthier, whitestudents to enroll for kindergarten and switch topublic school for first grade. This “bubble” in kin-dergarten enrollment could make an analysis of thatgrade unrepresentative of public and private schoolsmore generally.

Research directly on the racial integration impact ofvouchers in Milwaukee and Cleveland suggests thatthose programs contributed to greater opportunitiesfor racial integration. Fuller and Greiveldinger (2002)found that Milwaukee’s voucher program allowedparticipating students to attend more racially inte-grated private schools than could be found in theirprevious Milwaukee public schools. Greene (1999)found that 19% of students using a voucher to at-tend a private school in Cleveland went to a raciallyintegrated school, compared with only 5% of stu-dents in Cleveland’s public schools.

While researchers have not fully resolved theirdebates over the most appropriate methods formeasuring school integration and while the evi-dence on the effects of vouchers on integrationis far from definitive, the bulk of the researchusing reasonable methods suggests that ex-panded school choice contributes to higher lev-els of integration in school.

Evaluating the Effect of Vouchers on D.C.Public Schools After One Year

Our strategy for measuring the effect of voucherson public school achievement in Washington, D.C.,is to use different measures of the physical proxim-ity of public schools to private schools participat-ing in the voucher program as a proxy for thecompetition faced by those public schools. In theory,schools that are geographically closer to competingprivate schools, or that have a larger number of com-peting private schools within a given radius, willbe more likely to lose students to the voucher pro-gram than schools whose students have fewer pri-vate school options nearby.

Using multivariate regression techniques, we triedto identify the relationship between these measuresof voucher competition and the achievement of stu-dents in D.C. public schools. Each year, students inD.C. public schools are administered the Stanford-9 math and reading tests in grades three, five, eight,and ten. We collected aggregate school level meanscaled scores on these tests for each public school inthe District for the 2003–04 and 2004–05 schoolyears—the year before and the first year after imple-mentation of the voucher program—using the Dis-trict of Columbia Public Schools official website.5

We then calculated the gains that each school madeduring this one-year period.

We obtained the geographical address of everypublic school in Washington, D.C., from the Core ofCommon Data, made available by the U.S.Department of Education.6 We also obtained thegeographical address of each private school that isparticipating in the voucher program.7 Using acommercial mapping software package, we thenmeasured the distance between each public schooland the nearest private school participating in thevoucher program that served students in the samegrade levels.8 As an alternative measure of

Education Working Paper 10 and SCDP-06-01

January 2006

Results and Discussion of the Effect of VoucherCompetition on Public School AcademicPerformance

The results of our evaluations suggest that after oneyear, the voucher program has had no significantimpact on the D.C. public schools, positive or nega-tive. None of our 16 regression analyses produces astatistically significant finding for the chosen mea-sure of voucher competition—either distance to thenearest participating voucher school or the numberof competing voucher schools within a one-mile ra-dius (see Table 1).

There are several factors that could explain the nullfinding of our evaluation. First, it is possible that oneyear is not long enough for voucher competition tohave any positive or negative effect on public schools.Most previous evaluations finding positive impactsfrom other voucher programs were conducted atleast a few years after the programs were imple-mented. With time, it is possible that D.C.’s voucherprogram will lead to improvements or deteriorationin the quality of public schools.

The null finding could also be explained by neces-sary limitations in this study’s empirical design.While using proximity to competing schools to mea-sure competitive pressure has proven a workabledesign in previous school choice systemic-effect stud-ies (see, for example, Greene and Forster 2002 andHoxby 2001), this might not be the best measure ofcompetition for a metropolis such as Washington,D.C. Public transportation is abundant in the Dis-trict of Columbia, which could make mileage differ-ences between schools quite manageable for parents.Thus, public schools in the District might not facemeasurably less competitive pressure from a voucherschool five miles away compared with a voucherschool only a half-mile away.

Some might argue that our null finding occurs becausewe fail to measure any systemic academic effect thatthe District’s many charter schools might have hadon public schools. As mentioned earlier, there arenumerous charter schools in the District, each of whichtheoretically should have a similar competitive effectas the one that we are evaluating from vouchers. How-ever, the purpose of this study is to evaluate any addi-tional systemic academic effect that vouchers mighthave on public school performance, compared with

6

competition, we also counted the number ofparticipating private schools within a one-mileradius of each public school.

On average, D.C. public schools were located 0.68miles from the nearest competing private school,with a minimum distance of 0.06 miles and amaximum distance of 3.43 miles. The averageDistrict public school also had 2.33 competingvoucher schools within a one-mile radius, with aminimum of zero and a maximum of eightschools. Of the 151 public schools in Washington,D.C., that serve grades that are administered theStanford-9 and for which we have complete test-score information, there were 27 public schoolsthat have no competing voucher schools within aone-mile radius.9

To account for demographic characteristics thatcould affect test-score performance, we also ob-tained information on the percentage of studentsin each school who are white and the percentageof students enrolled in the free or reduced-pricelunch program during the 2003–04 school year.10

In our analyses, we also controlled for eachschool’s baseline test-score level in the subject andgrade being evaluated. Unfortunately, we wereunable to control for the change in these demo-graphic characteristics because the informationavailable from the U.S. Department of Educationlags by at least one year.

We performed a series of Ordinary Least Squaresregressions to measure the impact of geographicallocation to voucher-participating private schools onpublic school achievement. We performed indepen-dent regressions for each strategy for measuringcompetition, grade, and subject tested—for example,one regression for the effect of competition on fifth-grade reading-test-score gains using the distance tothe nearest voucher school as the measure of com-petition and another regression for the effect of com-petition on eighth-grade reading-test-score gainsusing the number of voucher schools within one mileas the measure of competition, and so on. In total,we performed 16 OLS analyses (two strategies formeasuring competition, two subjects in each of fourgrades). In each evaluation, the dependent variablewas the test-score gain that a school made in a gradeand subject of interest on the Stanford-9 between2003–04 and 2004–05.

An Evaluation of the Effects of D.C.’s Voucher Program onPublic School Achievement and Racial Integration After One Year

January 2006 7

the status quo. Thus, including any charter schooleffect would be unnecessary in our analysis becausecharter schools are already part of the academic en-vironment. Similarly, we do not account for the mostwidely used form of school choice—that is, locatingone’s residence in the desired school district or at-tendance zone—because such residential schoolchoice operates with or without the implementationof vouchers.

Aside from methodological considerations, there arealso theoretical reasons that we might expect a nullfinding for systemic effects in D.C. Most important,as discussed previously in this paper, the SchoolChoice Incentive Act was designed so that the pub-lic school system would not be adversely affectedfinancially from the program. Proponents as well asopponents of vouchers cite decreasing revenues asthe driving force for the academic effect that vouch-ers would have on public schools. Other voucherprograms where research has found an academiceffect from vouchers on public schools have usuallytied substantial resources to the loss of students fromvouchers. It is reasonable to argue that the lack ofthis financial aspect of the program is the most likelycause of our null finding.

Of course, it is also possible that our null finding iscaused by a true absence of any significant effect ofexpanded school choice on public school performance.Further analyses over time using a variety of ap-proaches—in D.C. and elsewhere—may help resolvethese uncertainties about the real relationship betweenvouchers and student achievement in public schools.

It is important to emphasize that the results of thisanalysis find that the voucher program has neitherhelped nor harmed D.C. public school academicachievement after one year. Thus, at least after itsfirst year, the School Choice Incentive vouchers haveneither helped to improve public schools in the Dis-trict, as advocates suggested, nor harmed thoseschools, as opponents suggested.

Evaluating the Potential Effects of RacialIntegration

This paper evaluates whether the OpportunityScholarship Program has increased the opportunityfor students to attend less segregated schools. Weutilize two strategies to measure racial segregation.

We first measure the extent to which each school’sracial composition differs from the racial composi-tion of the school-age population in the surround-ing metropolitan population, as defined by theUnited States Census. The greater absolute valueof the difference between a school’s demographiccharacteristics and the demographic characteristicsof the surrounding metro area, the more raciallysegregated the school. We then computed theweighted average difference for D.C. public schoolsand private schools participating in the program tosee which sector was more likely to offer students aracially integrated school environment.

Another approach to analyzing racial segregationwas to compare the percentage of public andvoucher-participating private schools with enroll-ments that are greater than 90% or 95% racially ho-mogeneous. This evaluation sheds light on thepercentage of schools that have student populationsthat simply cannot be considered to be racially inte-grated under any reasonable standard, regardless ofthe surrounding population.

To evaluate the impact of the D.C. voucher programon opportunities for racial integration, we collectedinformation on the racial composition of each publicschool and each private school participating in thevoucher program. For public schools, we acquireddata using the Core of Common Data.11 For voucher-participating private schools, we utilized a datasetmade available by the Washington Scholarship Pro-gram.12 These datasets provided information on thenumber of students who were nonwhite in eachschool, which we converted into percentages.13 Thus,all our analyses focus on integration between whiteand minority students and do not offer informationabout integration between different minority groups.

To compare public schools and the surroundingmetro population, we utilized data from the U.S.Census. We downloaded information on the racialcharacteristics of the school-age population (aged 5–18) in the Washington, D.C./Virginia/MarylandUrbanized Area as defined by the census. This is thepopulation from which area schools could reason-ably draw students.14

We used census data to calculate the percentage ofthe metro area’s school-age population that wasnonwhite. Comparing white and nonwhite popu-

Education Working Paper 10 and SCDP-06-01

January 2006

lation instead of breaking out each racial categorywas the only analysis possible, given publicly avail-able data on the private schools participating in thevoucher program. While other analyses might havebeen informative, restricting our focus to white/nonwhite integration is reasonable because it coin-cides with the general public’s primary concernwith racial segregation.

For each public and private school, we then calcu-lated the absolute value of the difference between itsnonwhite population and the surrounding metroarea’s nonwhite population. Next, we took the aver-age difference between the nonwhite school and areapopulations in the District, weighted for each school’senrollment. Failing to account for each school’s en-rollment size would give unnecessary weight to thepercentage of students who are nonwhite in schoolswith particularly small enrollment populations.

We also created two sets of dummy variables for eachschool: one indicating whether 90% of its studentpopulation was either white or nonwhite; and an-other indicating whether 95% of its student popula-tion was either white or nonwhite. We thencalculated the percentage of schools that had enroll-ments that were racially homogeneous by these defi-nitions. We again weighted the analysis to accountfor the size of each school’s enrollment.

Results and Discussion of Evaluation of RacialIntegration in Public and Voucher Schools

Table 2 reports our findings. According to the UnitedStates Census data we collected, the population ofthe Washington, D.C., metro area is 57.1% nonwhite.The absolute value of the difference between the non-white school-age population in the metro area andthe weighted average nonwhite population in theschool was 39.5% for public schools in Washington,D.C., and 33.8% among voucher-participating pri-vate schools. The smaller difference for privateschools indicates that private schools on averagehave a racial composition that more closely approxi-mates the racial composition of the broader commu-nity in which they are located. Neither sector iswonderfully integrated, but the voucher schoolswere somewhat less segregated.

Table 2 also shows that a weighted average 85.4% ofthe District’s public schools have student populations

8

that are at least 90% racially homogeneous, and 84.4%of them have student populations that are at least 95%homogeneous. Among private schools participatingin D.C.’s voucher program, however, a weighted av-erage of about 47.3% have student populations thatare at least 90% racially homogeneous, and about42.8% are 95% or more racially homogeneous.

Figure 1 illustrates the distribution of DC publicand participating voucher school students by themix of white and non-white students in theirschools. Of all students attending public schoolsin the district, 85.1% are enrolled in schools thatare at least 91% non-white compared to 42.8% ofstudents attending participating private schools.No students in district public schools attendschools than are between 0 and 10% non-whitecompared to 4.5% of private schools. The figureprovides information for each decile, showing thatvery few DC public school students attend schoolsthat approximate the 57.1% non-white average inthe metro area, while private school students aresomewhat more likely to be enrolled in schoolswith a representative racial mix.

The results of our analysis indicate that the ChoiceIncentive Act vouchers offer the opportunity for stu-dents to leave more segregated public schools forless segregated private schools accepting vouchers.That is, voucher-accepting private schools have ra-cial populations that better resemble the racial com-position of the surrounding metro area and are lesslikely to have student populations that are raciallyhomogeneous.

This analysis is unable to measure the actual directimpact that the Opportunity Scholarship Programhas had on racial integration in Washington, D.C.Such an evaluation would require individual leveldata on students who use vouchers and whichschools they attend, which has not been made pub-licly available.

However, we can make some reasonable inferencesabout the effect of the program on racial integra-tion from our results. Of those students who used avoucher to attend a private school, 94% are AfricanAmerican (Wolf et al., 2005). When we consider thatthe vast majority of students using the vouchers arenonwhite and that private schools are more raciallyintegrated than the pubic schools that these chil-

An Evaluation of the Effects of D.C.’s Voucher Program onPublic School Achievement and Racial Integration After One Year

January 2006 9

dren are leaving by having lower concentrations ofAfrican-American students, it is clear that the pro-gram is likely reducing racial segregation in school-ing. Nonetheless, further empirical evidenceutilizing individual level enrollment data from thevoucher program could substantially add to ourknowledge of the program’s effect on racial inte-gration in the District.

It is important to emphasize that, on average, nei-ther public schools nor private schools in D.C. ap-pear to have achieved what most people wouldconsider racial integration. However, the questionbefore us is whether the voucher program contrib-utes to opportunities for integration. Our analysisindicates that, since public schools are more raciallysegregated than private schools in the area, the D.C.voucher program will allow students the opportu-nity to leave more racially homogeneous schools forless segregated schools, which should lead to lowerrates of segregation for both groups. Future researchwill be necessary to explore the dynamic effects of

the voucher program on the level of racial integra-tion offered in each sector.

Conclusion

This paper is the beginning of a long-term analysisof the effects of the Choice Incentive Act voucher pro-gram on public school achievement and on racialintegration. After one year, we find that the programhas likely improved racial integration in the area’sschools but that it has had no significant effect onpublic school performance.

While these findings are meaningful to gauge theeffectiveness of the program so far, it is important tokeep our results in context of the long-term evalua-tion to which they belong and the broader researchliterature to which they add. It is possible that thetrue effects of the program will substantially changein time. As with all public policies, it is the long-termeffects of this and other voucher programs that arethe most meaningful for their futures.

Education Working Paper 10 and SCDP-06-01

January 200610

REFERENCES

Belfield, C. R., and Levin, H. H. (2002). The Effects of Competition on Educational Outcomes: A Review of theEvidence. National Center for the Study of Privatization in Education.

Blair, J. P., and Staley, S. (1995). “Quality Competition and Public Schools: Further Evidence.” Economicsof Education Review, 14(2).

Borland, M. V., and Howson, R. M. (1992). “Students’ Academic Achievement and the Degree of MarketConcentration in Education.” Economics of Education Review, 11(1).

———. (1993). “On The Determination of the Critical Level of Market Concentration in Education.”Economics of Education Review, 12(2).

———. (1995). “Competition, Expenditures, and Student Performance in Mathematics: A Comment onCrouch et al.” Public Choice, 87.

Chakrabarti, R. (2005). “Impact of Voucher Design on Public School Performance: Evidence from Floridaand Milwaukee.” Harvard University.

Dee, T. S. (1998). “Competition and the Quality of Public Schools.” Economics of Education Review, 17(4).Figlio, D. N., and C. E. Rouse. (2005). Do Accountability and Voucher Threats Improve Low-Performing

Schools?” University of Florida, Princeton University, and NBER.Fuller, H. L., and Greiveldinger, D. (2002). “The Impact of School Choice on Racial Integration in

Milwaukee Private Schools.” American Education Reform Council manuscript.Greene, J. P. (1998). “Civic Values in Public and Private Schools.” In Learning from School Choice, ed. P.

Peterson and B. Hassel. Brookings Institution.

ENDNOTES

1. National Center for Education Statistics, Digest of Education Statistics 2004, Table 165.2. National Assessment of Educational Progress, NAEP Data Tool, http://nces.ed.gov/

nationsreportcard.3. Data compiled from the Build a Table option made available by the National Center for Education

Statistics, Core of Common Data.4. Title III—DC Choice Incentive Act of 2003, Sec. 302 (6). Text available online at http://

www.schoolchoiceinfo.org/data/facts/DC_Choice.pdf.5. http://silicon.k12.dc.us/apds/APDSSummaryReports.asp.6. National Center for Education Statistics, Core of Common Data. For this study, the authors utilized

the Build a Table option of the Core of Common Data (http://nces.ed.gov/ccd/bat).7. “School Directory, D.C. Opportunity Scholarship Program, 2004–05 School Year,” Washington Schol-

arship Program, June 2005.8. The software used for this study was Street Atlas USA, 2004, from Delorme.9. There were a total of 170 public schools in the dataset downloaded from the Core of Common

Data, but we only have information on voucher competition for 151 schools. The reason for is that 14 werelisted as “ungraded,” three schools served only pre-K and kindergarten, and we were unable to match testscores for two of the schools using data from the D.C. Department of Education website.

10. National Center for Education Statistics, Core of Common Data.11. Ibid.12. “School Directory, D.C. Opportunity Scholarship Program, 2004–05 School Year,” Washington

Scholarship Program, June 2005.13. For the purposes of this paper, “white” refers to non-Hispanic whites.14. United States Census, American Fact Finder,

An Evaluation of the Effects of D.C.’s Voucher Program onPublic School Achievement and Racial Integration After One Year

January 2006 11

———. (1999). “The Racial, Economic, and Religious Context of Parental Choice in Cleveland.” Paperpresented at the Association for Public Policy Analysis and Management meeting.

———. (2002). 2001 Education Freedom Index. Manhattan Institute Civic Report 24.———. (2005). “Choosing Integration.” In School Choice and Diversity: What the Evidence Says, ed. J. T.

Scott. Teachers College Press.———, and Forster, G. (2002). “Rising to the Challenge: The Effect of School Choice on Public Schools in

Milwaukee and San Antonio.” Manhattan Institute Civic Bulletin 27.———, and Forster, G. (2004). “The Teachability Index: Can Disadvantaged Students Learn?” Manhattan

Institute Education Working Paper 6.———, and Winters, M. A. (2004). “Competition Passes the Test.” Education Next, 4(3).Hammons, C. W. (2001). “The Effects of Town Tuitioning in Vermont and Maine.” Milton and Rose D.

Friedman Foundation.Hanushek, E. A., and Rivkin, S. G. (2002) “Does Public School Competition Affect Teacher Quality?” In

The Economics of School Choice, ed. C. Hoxby. University of Chicago Press.Hoxby, C. (1994). “Do Private Schools Provide Competition for Public Schools?” NBER Working Paper

4978.———. (2000). “Does Competition Benefit Students and Taxpayers?” The American Economic Review,

90(5).———. (2001). “The Rising Tide.” Education Next, 1(4).Jepson, C. (1999) “The Effects of Private School Competition on Student Achievement.” Unpublished

paper.Marlow, M. L. (1997). “Public Education Supply and Student Performance.” Applied Economics, 29.———. (1999). “Spending, School Structure, and Public Education Quality: Evidence from California.”

Economics of Education Review, 19(1).Ritter, G., Rush, A., and Rush, J. (2002). “How Might School Choice Affect Racial Integration in Schools?

New Evidence from the ECLS-K.” Georgetown Public Policy Review.Sander, W. (1999). “Private Schools and Public School Achievement.” The Journal of Human Resources, 34.Walberg, H. J. (1993). “Losing Local Control of Education: Cost and Quality Implications.” Heartland

Policy Study, 59.West, M. R., and Peterson, P. E. (2005). “The Efficacy of Choice Threats Within School Accountability

Systems: Results from Legislatively Induced Experiments.” Paper presented before the AnnualConference of the Royal Economic Society, University of Nottingham.

Wolf, P., Gutmann, B., Eissa, N., Puma, M., and Silverberg, M. (2005). “Evaluation of the DC OpportunityScholarship Program: First-Year Report on Participation.” U.S. Department of Education, NationalCenter for Education Evaluation and Regional Assistance. U.S. Government Printing Office.

Zanzig, B. R. (1997). “Measuring the Impact of Competition in Local Government Education Markets onthe Cognitive Achievement of Students.” Economics of Education Review, 16(4).

Education Working Paper 10 and SCDP-06-01

January 200612

APPENDIX

Table 2: Measures of School Segregation

D.C. Public Schools Voucher Schools

Difference in Percent Minority from Ideal Integration 39.5% 33.8%Percent of Schools 90% Homogeneous or More 85.4% 47.3%Percent of Schools 95% Homogeneous or More 84.4% 42.8%N 169 52

Table 1: Effect of Voucher Competition on Public School Performance

MATH READING

Number of Number ofDistance to Voucher Schools Distance to Voucher Schools

Nearest Voucher with Same Grade Nearest Voucher with Same GradeSchool with Level Within School with Level Within

Same Grade Level* One-Mile Radius Same Grade Level* One-Mile Radius

Grade 3

Unstandardized Coefficient -0.456 -0.344 -0.772 -0.45Standard Error 4.59 1.05 4.232 0.965N 104 104 103 103

Grade 5

Unstandardized Coefficient -0.655 1.186 1.95 0.172Standard Error 3.285 0.746 2.865 0.668N 103 103 103 103

Grade 8

Unstandardized Coefficient -3.982 0.905 -5.587 1.545Standard Error 2.45 0.655 4.717 1.244N 26 26 26 26

Grade 10

Unstandardized Coefficient -0.177 -1.591 0.019 3.579Standard Error 2.611 2.704 2.592 2.682N 19 19 19 19

(controlling for percentage of students who are white, percentage of students enrolled in the free or reduced-price lunch program,and baseline 2004 test score in same grade and subject)

*In the distance analyses, a negative coefficient indicates that public schools nearer to the closest voucher school make the largestgains. Thus, a negative coefficient is most consistent with the hypothesis that voucher competition has improved public schoolperformance in Washington, D.C.

An Evaluation of the Effects of D.C.’s Voucher Program onPublic School Achievement and Racial Integration After One Year

January 2006 13

Figure 1: Percent Minority Enrollment in DecilesP

erce

ntag

e o

f St

uden

ts

Percentage Minority Enrollment

EXECUTIVE DIRECTOR

Henry Olsen

ADVISORY BOARD

Stephen Goldsmith, Chairman EmeritasMayor Jerry BrownMayor Manuel A.DiazMayor Martin O’MalleyMayor Rick Baker

FELLOWS

Jay P. GreeneGeorge L. Kelling

Edmund J. McMahonPeter D. Salins

The Center for Civic Innovation’s (CCI) purpose is to improve the quality of life in cities by shaping publicpolicy and enriching public discourse on urban issues.

CCI sponsors the publication of books like The Entrepreneurial City: A How-To Handbook for Urban Innovators,which contains brief essays from America’s leading mayors explaining how they improved their cities’ qualityof life; Stephen Goldsmith’s The Twenty-First Century City, which provides a blueprint for getting America’scities back in shape; and George Kelling’s and Catherine Coles’ Fixing Broken Windows, which explores thetheory widely credited with reducing the rate of crime in New York and other cities. CCI also hosts conferences,publishes studies, and holds luncheon forums where prominent local and national leaders are givenopportunities to present their views on critical urban issues. Cities on a Hill, CCI’s newsletter, highlights theongoing work of innovative mayors across the country.

The Manhattan Institute is a 501(C)(3) nonprofit organization. Contributions are tax-deductible to the fullestextent of the law. EIN #13-2912529

M A N H A T T A N I N S T I T U T E F O R P O L I C Y R E S E A R C H

M52 Vanderbilt Avenue • New York, NY 10017

www.manhattan-institute.org

INon-Profit

OrganizationUS Postage

PAIDPermit 04001

New York, NY